pycaret.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "initial_id",
   "metadata": {
    "collapsed": true,
    "ExecuteTime": {
     "end_time": "2024-02-05T20:06:40.704406407Z",
     "start_time": "2024-02-05T20:06:34.677121265Z"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from pycaret.classification import *\n",
    "\n",
    "# Read data\n",
    "train = pd.read_csv('dataset/train.csv')\n",
    "test = pd.read_csv('dataset/test.csv')\n",
    "submit = pd.read_csv('dataset/submit.csv')\n",
    "\n",
    "# Merge test and submit\n",
    "test_df = pd.merge(test, submit, on='id', how='left')\n",
    "\n",
    "# Filling NaN values in specific columns\n",
    "train['title'].fillna('Untitled', inplace=True)\n",
    "train['author'].fillna('unknown', inplace=True)\n",
    "\n",
    "# Dropping rows with NaN values in the 'text' column\n",
    "train.dropna(subset=['text'], inplace=True)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "outputs": [],
   "source": [
    "# Combine text features\n",
    "train['combined'] = train['title'].astype(str) + ' ' + train['author'].astype(str) + ' ' + train['text'].astype(str)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-02-05T20:06:40.829859715Z",
     "start_time": "2024-02-05T20:06:40.827024795Z"
    }
   },
   "id": "8138d3cbf7f42afa"
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001B[33mWARNING: Skipping /mnt/Hossein-HDD/Projects/Fake-news-detection/venv/lib/python3.11/site-packages/matplotlib-3.8.2.dist-info due to invalid metadata entry 'name'\u001B[0m\u001B[33m\r\n",
      "\u001B[0m\u001B[33mWARNING: Skipping /mnt/Hossein-HDD/Projects/Fake-news-detection/venv/lib/python3.11/site-packages/matplotlib-3.8.2.dist-info due to invalid metadata entry 'name'\u001B[0m\u001B[33m\r\n",
      "\u001B[0mRequirement already satisfied: mlflow in ./venv/lib/python3.11/site-packages (2.9.2)\r\n",
      "Requirement already satisfied: click<9,>=7.0 in ./venv/lib/python3.11/site-packages (from mlflow) (8.1.7)\r\n",
      "Requirement already satisfied: cloudpickle<4 in ./venv/lib/python3.11/site-packages (from mlflow) (3.0.0)\r\n",
      "Requirement already satisfied: databricks-cli<1,>=0.8.7 in ./venv/lib/python3.11/site-packages (from mlflow) (0.18.0)\r\n",
      "Requirement already satisfied: entrypoints<1 in ./venv/lib/python3.11/site-packages (from mlflow) (0.4)\r\n",
      "Requirement already satisfied: gitpython<4,>=2.1.0 in ./venv/lib/python3.11/site-packages (from mlflow) (3.1.41)\r\n",
      "Requirement already satisfied: pyyaml<7,>=5.1 in ./venv/lib/python3.11/site-packages (from mlflow) (6.0.1)\r\n",
      "Requirement already satisfied: protobuf<5,>=3.12.0 in ./venv/lib/python3.11/site-packages (from mlflow) (4.25.2)\r\n",
      "Requirement already satisfied: pytz<2024 in ./venv/lib/python3.11/site-packages (from mlflow) (2023.3.post1)\r\n",
      "Requirement already satisfied: requests<3,>=2.17.3 in ./venv/lib/python3.11/site-packages (from mlflow) (2.31.0)\r\n",
      "Requirement already satisfied: packaging<24 in ./venv/lib/python3.11/site-packages (from mlflow) (23.2)\r\n",
      "Requirement already satisfied: importlib-metadata!=4.7.0,<8,>=3.7.0 in ./venv/lib/python3.11/site-packages (from mlflow) (7.0.1)\r\n",
      "Requirement already satisfied: sqlparse<1,>=0.4.0 in ./venv/lib/python3.11/site-packages (from mlflow) (0.4.4)\r\n",
      "Requirement already satisfied: alembic!=1.10.0,<2 in ./venv/lib/python3.11/site-packages (from mlflow) (1.13.1)\r\n",
      "Requirement already satisfied: docker<7,>=4.0.0 in ./venv/lib/python3.11/site-packages (from mlflow) (6.1.3)\r\n",
      "Requirement already satisfied: Flask<4 in ./venv/lib/python3.11/site-packages (from mlflow) (3.0.1)\r\n",
      "Requirement already satisfied: numpy<2 in ./venv/lib/python3.11/site-packages (from mlflow) (1.25.2)\r\n",
      "Requirement already satisfied: scipy<2 in ./venv/lib/python3.11/site-packages (from mlflow) (1.10.1)\r\n",
      "Requirement already satisfied: pandas<3 in ./venv/lib/python3.11/site-packages (from mlflow) (1.5.3)\r\n",
      "Requirement already satisfied: querystring-parser<2 in ./venv/lib/python3.11/site-packages (from mlflow) (1.2.4)\r\n",
      "Requirement already satisfied: sqlalchemy<3,>=1.4.0 in ./venv/lib/python3.11/site-packages (from mlflow) (2.0.25)\r\n",
      "Requirement already satisfied: scikit-learn<2 in ./venv/lib/python3.11/site-packages (from mlflow) (1.2.2)\r\n",
      "Requirement already satisfied: pyarrow<15,>=4.0.0 in ./venv/lib/python3.11/site-packages (from mlflow) (14.0.2)\r\n",
      "Requirement already satisfied: markdown<4,>=3.3 in ./venv/lib/python3.11/site-packages (from mlflow) (3.5.2)\r\n",
      "Requirement already satisfied: matplotlib<4 in ./venv/lib/python3.11/site-packages (from mlflow) (3.6.0)\r\n",
      "Requirement already satisfied: gunicorn<22 in ./venv/lib/python3.11/site-packages (from mlflow) (21.2.0)\r\n",
      "Requirement already satisfied: Jinja2<4,>=2.11 in ./venv/lib/python3.11/site-packages (from mlflow) (3.1.3)\r\n",
      "Requirement already satisfied: Mako in ./venv/lib/python3.11/site-packages (from alembic!=1.10.0,<2->mlflow) (1.3.1)\r\n",
      "Requirement already satisfied: typing-extensions>=4 in ./venv/lib/python3.11/site-packages (from alembic!=1.10.0,<2->mlflow) (4.9.0)\r\n",
      "Requirement already satisfied: pyjwt>=1.7.0 in ./venv/lib/python3.11/site-packages (from databricks-cli<1,>=0.8.7->mlflow) (2.8.0)\r\n",
      "Requirement already satisfied: oauthlib>=3.1.0 in ./venv/lib/python3.11/site-packages (from databricks-cli<1,>=0.8.7->mlflow) (3.2.2)\r\n",
      "Requirement already satisfied: tabulate>=0.7.7 in ./venv/lib/python3.11/site-packages (from databricks-cli<1,>=0.8.7->mlflow) (0.9.0)\r\n",
      "Requirement already satisfied: six>=1.10.0 in ./venv/lib/python3.11/site-packages (from databricks-cli<1,>=0.8.7->mlflow) (1.16.0)\r\n",
      "Requirement already satisfied: urllib3<3,>=1.26.7 in ./venv/lib/python3.11/site-packages (from databricks-cli<1,>=0.8.7->mlflow) (2.1.0)\r\n",
      "Requirement already satisfied: websocket-client>=0.32.0 in ./venv/lib/python3.11/site-packages (from docker<7,>=4.0.0->mlflow) (1.7.0)\r\n",
      "Requirement already satisfied: Werkzeug>=3.0.0 in ./venv/lib/python3.11/site-packages (from Flask<4->mlflow) (3.0.1)\r\n",
      "Requirement already satisfied: itsdangerous>=2.1.2 in ./venv/lib/python3.11/site-packages (from Flask<4->mlflow) (2.1.2)\r\n",
      "Requirement already satisfied: blinker>=1.6.2 in ./venv/lib/python3.11/site-packages (from Flask<4->mlflow) (1.7.0)\r\n",
      "Requirement already satisfied: gitdb<5,>=4.0.1 in ./venv/lib/python3.11/site-packages (from gitpython<4,>=2.1.0->mlflow) (4.0.11)\r\n",
      "Requirement already satisfied: zipp>=0.5 in ./venv/lib/python3.11/site-packages (from importlib-metadata!=4.7.0,<8,>=3.7.0->mlflow) (3.17.0)\r\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in ./venv/lib/python3.11/site-packages (from Jinja2<4,>=2.11->mlflow) (2.1.4)\r\n",
      "Requirement already satisfied: contourpy>=1.0.1 in ./venv/lib/python3.11/site-packages (from matplotlib<4->mlflow) (1.2.0)\r\n",
      "Requirement already satisfied: cycler>=0.10 in ./venv/lib/python3.11/site-packages (from matplotlib<4->mlflow) (0.12.1)\r\n",
      "Requirement already satisfied: fonttools>=4.22.0 in ./venv/lib/python3.11/site-packages (from matplotlib<4->mlflow) (4.47.2)\r\n",
      "Requirement already satisfied: kiwisolver>=1.0.1 in ./venv/lib/python3.11/site-packages (from matplotlib<4->mlflow) (1.4.5)\r\n",
      "Requirement already satisfied: pillow>=6.2.0 in ./venv/lib/python3.11/site-packages (from matplotlib<4->mlflow) (10.2.0)\r\n",
      "Requirement already satisfied: pyparsing>=2.2.1 in ./venv/lib/python3.11/site-packages (from matplotlib<4->mlflow) (3.1.1)\r\n",
      "Requirement already satisfied: python-dateutil>=2.7 in ./venv/lib/python3.11/site-packages (from matplotlib<4->mlflow) (2.8.2)\r\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in ./venv/lib/python3.11/site-packages (from requests<3,>=2.17.3->mlflow) (3.3.2)\r\n",
      "Requirement already satisfied: idna<4,>=2.5 in ./venv/lib/python3.11/site-packages (from requests<3,>=2.17.3->mlflow) (3.6)\r\n",
      "Requirement already satisfied: certifi>=2017.4.17 in ./venv/lib/python3.11/site-packages (from requests<3,>=2.17.3->mlflow) (2023.11.17)\r\n",
      "Requirement already satisfied: joblib>=1.1.1 in ./venv/lib/python3.11/site-packages (from scikit-learn<2->mlflow) (1.3.2)\r\n",
      "Requirement already satisfied: threadpoolctl>=2.0.0 in ./venv/lib/python3.11/site-packages (from scikit-learn<2->mlflow) (3.2.0)\r\n",
      "Requirement already satisfied: greenlet!=0.4.17 in ./venv/lib/python3.11/site-packages (from sqlalchemy<3,>=1.4.0->mlflow) (3.0.3)\r\n",
      "Requirement already satisfied: smmap<6,>=3.0.1 in ./venv/lib/python3.11/site-packages (from gitdb<5,>=4.0.1->gitpython<4,>=2.1.0->mlflow) (5.0.1)\r\n",
      "\u001B[33mWARNING: Skipping /mnt/Hossein-HDD/Projects/Fake-news-detection/venv/lib/python3.11/site-packages/matplotlib-3.8.2.dist-info due to invalid metadata entry 'name'\u001B[0m\u001B[33m\r\n",
      "\u001B[0m\u001B[33mWARNING: Skipping /mnt/Hossein-HDD/Projects/Fake-news-detection/venv/lib/python3.11/site-packages/matplotlib-3.8.2.dist-info due to invalid metadata entry 'name'\u001B[0m\u001B[33m\r\n",
      "\u001B[0m\u001B[33mWARNING: Skipping /mnt/Hossein-HDD/Projects/Fake-news-detection/venv/lib/python3.11/site-packages/matplotlib-3.8.2.dist-info due to invalid metadata entry 'name'\u001B[0m\u001B[33m\r\n",
      "\u001B[0m\u001B[33mWARNING: Skipping /mnt/Hossein-HDD/Projects/Fake-news-detection/venv/lib/python3.11/site-packages/matplotlib-3.8.2.dist-info due to invalid metadata entry 'name'\u001B[0m\u001B[33m\r\n",
      "\u001B[0m\r\n",
      "\u001B[1m[\u001B[0m\u001B[34;49mnotice\u001B[0m\u001B[1;39;49m]\u001B[0m\u001B[39;49m A new release of pip is available: \u001B[0m\u001B[31;49m23.2.1\u001B[0m\u001B[39;49m -> \u001B[0m\u001B[32;49m24.0\u001B[0m\r\n",
      "\u001B[1m[\u001B[0m\u001B[34;49mnotice\u001B[0m\u001B[1;39;49m]\u001B[0m\u001B[39;49m To update, run: \u001B[0m\u001B[32;49mpip install --upgrade pip\u001B[0m\r\n"
     ]
    },
    {
     "data": {
      "text/plain": "<IPython.core.display.HTML object>",
      "text/html": ""
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024/02/05 23:37:04 INFO mlflow.tracking.fluent: Experiment with name 'rf_experiment' does not exist. Creating a new experiment.\n"
     ]
    },
    {
     "data": {
      "text/plain": "                                                                    \n                                                                    \nInitiated  . . . . . . . . . . . . . . . . . .              23:37:07\nStatus     . . . . . . . . . . . . . . . . . .  Loading Dependencies\nEstimator  . . . . . . . . . . . . . . . . . .     Compiling Library",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n    <tr>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>Initiated</th>\n      <td>. . . . . . . . . . . . . . . . . .</td>\n      <td>23:37:07</td>\n    </tr>\n    <tr>\n      <th>Status</th>\n      <td>. . . . . . . . . . . . . . . . . .</td>\n      <td>Loading Dependencies</td>\n    </tr>\n    <tr>\n      <th>Estimator</th>\n      <td>. . . . . . . . . . . . . . . . . .</td>\n      <td>Compiling Library</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<IPython.core.display.HTML object>",
      "text/html": ""
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "Processing:   0%|          | 0/65 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "86cd4d1e9ee94c73ab51633e3802dac6"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "                                                                    \n                                                                    \nInitiated  . . . . . . . . . . . . . . . . . .              23:43:16\nStatus     . . . . . . . . . . . . . . . . . .  Loading Dependencies\nEstimator  . . . . . . . . . . . . . . . . . .     Compiling Library",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n    <tr>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>Initiated</th>\n      <td>. . . . . . . . . . . . . . . . . .</td>\n      <td>23:43:16</td>\n    </tr>\n    <tr>\n      <th>Status</th>\n      <td>. . . . . . . . . . . . . . . . . .</td>\n      <td>Loading Dependencies</td>\n    </tr>\n    <tr>\n      <th>Estimator</th>\n      <td>. . . . . . . . . . . . . . . . . .</td>\n      <td>Compiling Library</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<IPython.core.display.HTML object>",
      "text/html": ""
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "Processing:   0%|          | 0/4 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "e41cad6dc79a434596dcffb4c5b7df1a"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "                                                                    \n                                                                    \nInitiated  . . . . . . . . . . . . . . . . . .              23:43:28\nStatus     . . . . . . . . . . . . . . . . . .  Loading Dependencies\nEstimator  . . . . . . . . . . . . . . . . . .     Compiling Library",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n    <tr>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>Initiated</th>\n      <td>. . . . . . . . . . . . . . . . . .</td>\n      <td>23:43:28</td>\n    </tr>\n    <tr>\n      <th>Status</th>\n      <td>. . . . . . . . . . . . . . . . . .</td>\n      <td>Loading Dependencies</td>\n    </tr>\n    <tr>\n      <th>Estimator</th>\n      <td>. . . . . . . . . . . . . . . . . .</td>\n      <td>Compiling Library</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<IPython.core.display.HTML object>",
      "text/html": ""
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "Processing:   0%|          | 0/7 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "20c597e46c434589907d76341c9bff13"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 10 folds for each of 10 candidates, totalling 100 fits\n",
      "Original model was better than the tuned model, hence it will be returned. NOTE: The display metrics are for the tuned model (not the original one).\n"
     ]
    }
   ],
   "source": [
    "! pip install mlflow\n",
    "# Set up PyCaret\n",
    "clf1 = setup(data=train, target='label', session_id=42, log_experiment=True, experiment_name='rf_experiment')\n",
    "\n",
    "# Compare models and select the best\n",
    "best_model = compare_models()\n",
    "\n",
    "# Create and train Random Forest model\n",
    "rf = create_model('rf')\n",
    "\n",
    "# Tune the Random Forest model\n",
    "tuned_rf = tune_model(rf)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-02-05T20:15:02.416656549Z",
     "start_time": "2024-02-05T20:06:40.833267396Z"
    }
   },
   "id": "2f36bbecfdcc6463"
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "outputs": [
    {
     "data": {
      "text/plain": "interactive(children=(ToggleButtons(description='Plot Type:', icons=('',), options=(('Pipeline Plot', 'pipelin…",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "3226255e5321446bb9c0cc1c547b003e"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<IPython.core.display.HTML object>",
      "text/html": ""
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "<Figure size 800x550 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Transformation Pipeline and Model Successfully Saved\n"
     ]
    },
    {
     "data": {
      "text/plain": "(Pipeline(memory=Memory(location=None),\n          steps=[('numerical_imputer',\n                  TransformerWrapper(exclude=None, include=['id'],\n                                     transformer=SimpleImputer(add_indicator=False,\n                                                               copy=True,\n                                                               fill_value=None,\n                                                               keep_empty_features=False,\n                                                               missing_values=nan,\n                                                               strategy='mean',\n                                                               verbose='deprecated'))),\n                 ('categorical_imputer',\n                  TransformerWrapper(exclude=None,\n                                     include=['title', 'author'...\n                  RandomForestClassifier(bootstrap=True, ccp_alpha=0.0,\n                                         class_weight=None, criterion='gini',\n                                         max_depth=None, max_features='sqrt',\n                                         max_leaf_nodes=None, max_samples=None,\n                                         min_impurity_decrease=0.0,\n                                         min_samples_leaf=1, min_samples_split=2,\n                                         min_weight_fraction_leaf=0.0,\n                                         n_estimators=100, n_jobs=-1,\n                                         oob_score=False, random_state=42,\n                                         verbose=0, warm_start=False))],\n          verbose=False),\n 'random_forest_model.pkl')"
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Evaluate the tuned model\n",
    "evaluate_model(tuned_rf)\n",
    "\n",
    "# Plot confusion matrix\n",
    "plot_model(tuned_rf, plot='confusion_matrix')\n",
    "\n",
    "# Save the model\n",
    "save_model(tuned_rf, 'random_forest_model')"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-02-05T20:15:10.662722887Z",
     "start_time": "2024-02-05T20:15:02.421221332Z"
    }
   },
   "id": "23cb3df849d38751"
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "outputs": [
    {
     "data": {
      "text/plain": "<IPython.core.display.HTML object>",
      "text/html": ""
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "KeyError",
     "evalue": "\"['combined'] not in index\"",
     "output_type": "error",
     "traceback": [
      "\u001B[0;31m---------------------------------------------------------------------------\u001B[0m",
      "\u001B[0;31mKeyError\u001B[0m                                  Traceback (most recent call last)",
      "Cell \u001B[0;32mIn[5], line 5\u001B[0m\n\u001B[1;32m      2\u001B[0m final_model \u001B[38;5;241m=\u001B[39m finalize_model(tuned_rf)\n\u001B[1;32m      4\u001B[0m \u001B[38;5;66;03m# Predictions on the test set\u001B[39;00m\n\u001B[0;32m----> 5\u001B[0m test_predictions \u001B[38;5;241m=\u001B[39m \u001B[43mpredict_model\u001B[49m\u001B[43m(\u001B[49m\u001B[43mtuned_rf\u001B[49m\u001B[43m,\u001B[49m\u001B[43m \u001B[49m\u001B[43mdata\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mtest\u001B[49m\u001B[43m)\u001B[49m\n\u001B[1;32m      7\u001B[0m \u001B[38;5;66;03m# Display the predictions\u001B[39;00m\n\u001B[1;32m      8\u001B[0m \u001B[38;5;28mprint\u001B[39m(test_predictions\u001B[38;5;241m.\u001B[39mhead())\n",
      "File \u001B[0;32m/mnt/Hossein-HDD/Projects/Fake-news-detection/venv/lib/python3.11/site-packages/pycaret/classification/functional.py:2166\u001B[0m, in \u001B[0;36mpredict_model\u001B[0;34m(estimator, data, probability_threshold, encoded_labels, raw_score, round, verbose)\u001B[0m\n\u001B[1;32m   2163\u001B[0m \u001B[38;5;28;01mif\u001B[39;00m experiment \u001B[38;5;129;01mis\u001B[39;00m \u001B[38;5;28;01mNone\u001B[39;00m:\n\u001B[1;32m   2164\u001B[0m     experiment \u001B[38;5;241m=\u001B[39m _EXPERIMENT_CLASS()\n\u001B[0;32m-> 2166\u001B[0m \u001B[38;5;28;01mreturn\u001B[39;00m \u001B[43mexperiment\u001B[49m\u001B[38;5;241;43m.\u001B[39;49m\u001B[43mpredict_model\u001B[49m\u001B[43m(\u001B[49m\n\u001B[1;32m   2167\u001B[0m \u001B[43m    \u001B[49m\u001B[43mestimator\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mestimator\u001B[49m\u001B[43m,\u001B[49m\n\u001B[1;32m   2168\u001B[0m \u001B[43m    \u001B[49m\u001B[43mdata\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mdata\u001B[49m\u001B[43m,\u001B[49m\n\u001B[1;32m   2169\u001B[0m \u001B[43m    \u001B[49m\u001B[43mprobability_threshold\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mprobability_threshold\u001B[49m\u001B[43m,\u001B[49m\n\u001B[1;32m   2170\u001B[0m \u001B[43m    \u001B[49m\u001B[43mencoded_labels\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mencoded_labels\u001B[49m\u001B[43m,\u001B[49m\n\u001B[1;32m   2171\u001B[0m \u001B[43m    \u001B[49m\u001B[43mraw_score\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mraw_score\u001B[49m\u001B[43m,\u001B[49m\n\u001B[1;32m   2172\u001B[0m \u001B[43m    \u001B[49m\u001B[38;5;28;43mround\u001B[39;49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[38;5;28;43mround\u001B[39;49m\u001B[43m,\u001B[49m\n\u001B[1;32m   2173\u001B[0m \u001B[43m    \u001B[49m\u001B[43mverbose\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mverbose\u001B[49m\u001B[43m,\u001B[49m\n\u001B[1;32m   2174\u001B[0m \u001B[43m\u001B[49m\u001B[43m)\u001B[49m\n",
      "File \u001B[0;32m/mnt/Hossein-HDD/Projects/Fake-news-detection/venv/lib/python3.11/site-packages/pycaret/classification/oop.py:2813\u001B[0m, in \u001B[0;36mClassificationExperiment.predict_model\u001B[0;34m(self, estimator, data, probability_threshold, encoded_labels, raw_score, round, verbose)\u001B[0m\n\u001B[1;32m   2741\u001B[0m \u001B[38;5;28;01mdef\u001B[39;00m \u001B[38;5;21mpredict_model\u001B[39m(\n\u001B[1;32m   2742\u001B[0m     \u001B[38;5;28mself\u001B[39m,\n\u001B[1;32m   2743\u001B[0m     estimator,\n\u001B[0;32m   (...)\u001B[0m\n\u001B[1;32m   2749\u001B[0m     verbose: \u001B[38;5;28mbool\u001B[39m \u001B[38;5;241m=\u001B[39m \u001B[38;5;28;01mTrue\u001B[39;00m,\n\u001B[1;32m   2750\u001B[0m ) \u001B[38;5;241m-\u001B[39m\u001B[38;5;241m>\u001B[39m pd\u001B[38;5;241m.\u001B[39mDataFrame:\n\u001B[1;32m   2751\u001B[0m \u001B[38;5;250m    \u001B[39m\u001B[38;5;124;03m\"\"\"\u001B[39;00m\n\u001B[1;32m   2752\u001B[0m \u001B[38;5;124;03m    This function predicts ``Label`` and ``Score`` (probability of predicted\u001B[39;00m\n\u001B[1;32m   2753\u001B[0m \u001B[38;5;124;03m    class) using a trained model. When ``data`` is None, it predicts label and\u001B[39;00m\n\u001B[0;32m   (...)\u001B[0m\n\u001B[1;32m   2810\u001B[0m \n\u001B[1;32m   2811\u001B[0m \u001B[38;5;124;03m    \"\"\"\u001B[39;00m\n\u001B[0;32m-> 2813\u001B[0m     \u001B[38;5;28;01mreturn\u001B[39;00m \u001B[38;5;28;43msuper\u001B[39;49m\u001B[43m(\u001B[49m\u001B[43m)\u001B[49m\u001B[38;5;241;43m.\u001B[39;49m\u001B[43mpredict_model\u001B[49m\u001B[43m(\u001B[49m\n\u001B[1;32m   2814\u001B[0m \u001B[43m        \u001B[49m\u001B[43mestimator\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mestimator\u001B[49m\u001B[43m,\u001B[49m\n\u001B[1;32m   2815\u001B[0m \u001B[43m        \u001B[49m\u001B[43mdata\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mdata\u001B[49m\u001B[43m,\u001B[49m\n\u001B[1;32m   2816\u001B[0m \u001B[43m        \u001B[49m\u001B[43mprobability_threshold\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mprobability_threshold\u001B[49m\u001B[43m,\u001B[49m\n\u001B[1;32m   2817\u001B[0m \u001B[43m        \u001B[49m\u001B[43mencoded_labels\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mencoded_labels\u001B[49m\u001B[43m,\u001B[49m\n\u001B[1;32m   2818\u001B[0m \u001B[43m        \u001B[49m\u001B[43mraw_score\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mraw_score\u001B[49m\u001B[43m,\u001B[49m\n\u001B[1;32m   2819\u001B[0m \u001B[43m        \u001B[49m\u001B[38;5;28;43mround\u001B[39;49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[38;5;28;43mround\u001B[39;49m\u001B[43m,\u001B[49m\n\u001B[1;32m   2820\u001B[0m \u001B[43m        \u001B[49m\u001B[43mverbose\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mverbose\u001B[49m\u001B[43m,\u001B[49m\n\u001B[1;32m   2821\u001B[0m \u001B[43m    \u001B[49m\u001B[43m)\u001B[49m\n",
      "File \u001B[0;32m/mnt/Hossein-HDD/Projects/Fake-news-detection/venv/lib/python3.11/site-packages/pycaret/internal/pycaret_experiment/supervised_experiment.py:4952\u001B[0m, in \u001B[0;36m_SupervisedExperiment.predict_model\u001B[0;34m(self, estimator, data, probability_threshold, encoded_labels, raw_score, round, verbose, ml_usecase, preprocess)\u001B[0m\n\u001B[1;32m   4950\u001B[0m X_test_untransformed \u001B[38;5;241m=\u001B[39m data\n\u001B[1;32m   4951\u001B[0m y_test_untransformed \u001B[38;5;241m=\u001B[39m target\n\u001B[0;32m-> 4952\u001B[0m data \u001B[38;5;241m=\u001B[39m \u001B[43mdata\u001B[49m\u001B[43m[\u001B[49m\u001B[43mX_columns\u001B[49m\u001B[43m]\u001B[49m  \u001B[38;5;66;03m# Ignore all columns but the originals\u001B[39;00m\n\u001B[1;32m   4953\u001B[0m \u001B[38;5;28;01mif\u001B[39;00m preprocess:\n\u001B[1;32m   4954\u001B[0m     X_test_ \u001B[38;5;241m=\u001B[39m pipeline\u001B[38;5;241m.\u001B[39mtransform(\n\u001B[1;32m   4955\u001B[0m         X\u001B[38;5;241m=\u001B[39mdata,\n\u001B[1;32m   4956\u001B[0m         y\u001B[38;5;241m=\u001B[39m(target \u001B[38;5;28;01mif\u001B[39;00m preprocess \u001B[38;5;241m!=\u001B[39m \u001B[38;5;124m\"\u001B[39m\u001B[38;5;124mfeatures\u001B[39m\u001B[38;5;124m\"\u001B[39m \u001B[38;5;28;01melse\u001B[39;00m \u001B[38;5;28;01mNone\u001B[39;00m),\n\u001B[1;32m   4957\u001B[0m     )\n",
      "File \u001B[0;32m/mnt/Hossein-HDD/Projects/Fake-news-detection/venv/lib/python3.11/site-packages/pandas/core/frame.py:3813\u001B[0m, in \u001B[0;36mDataFrame.__getitem__\u001B[0;34m(self, key)\u001B[0m\n\u001B[1;32m   3811\u001B[0m     \u001B[38;5;28;01mif\u001B[39;00m is_iterator(key):\n\u001B[1;32m   3812\u001B[0m         key \u001B[38;5;241m=\u001B[39m \u001B[38;5;28mlist\u001B[39m(key)\n\u001B[0;32m-> 3813\u001B[0m     indexer \u001B[38;5;241m=\u001B[39m \u001B[38;5;28;43mself\u001B[39;49m\u001B[38;5;241;43m.\u001B[39;49m\u001B[43mcolumns\u001B[49m\u001B[38;5;241;43m.\u001B[39;49m\u001B[43m_get_indexer_strict\u001B[49m\u001B[43m(\u001B[49m\u001B[43mkey\u001B[49m\u001B[43m,\u001B[49m\u001B[43m \u001B[49m\u001B[38;5;124;43m\"\u001B[39;49m\u001B[38;5;124;43mcolumns\u001B[39;49m\u001B[38;5;124;43m\"\u001B[39;49m\u001B[43m)\u001B[49m[\u001B[38;5;241m1\u001B[39m]\n\u001B[1;32m   3815\u001B[0m \u001B[38;5;66;03m# take() does not accept boolean indexers\u001B[39;00m\n\u001B[1;32m   3816\u001B[0m \u001B[38;5;28;01mif\u001B[39;00m \u001B[38;5;28mgetattr\u001B[39m(indexer, \u001B[38;5;124m\"\u001B[39m\u001B[38;5;124mdtype\u001B[39m\u001B[38;5;124m\"\u001B[39m, \u001B[38;5;28;01mNone\u001B[39;00m) \u001B[38;5;241m==\u001B[39m \u001B[38;5;28mbool\u001B[39m:\n",
      "File \u001B[0;32m/mnt/Hossein-HDD/Projects/Fake-news-detection/venv/lib/python3.11/site-packages/pandas/core/indexes/base.py:6070\u001B[0m, in \u001B[0;36mIndex._get_indexer_strict\u001B[0;34m(self, key, axis_name)\u001B[0m\n\u001B[1;32m   6067\u001B[0m \u001B[38;5;28;01melse\u001B[39;00m:\n\u001B[1;32m   6068\u001B[0m     keyarr, indexer, new_indexer \u001B[38;5;241m=\u001B[39m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39m_reindex_non_unique(keyarr)\n\u001B[0;32m-> 6070\u001B[0m \u001B[38;5;28;43mself\u001B[39;49m\u001B[38;5;241;43m.\u001B[39;49m\u001B[43m_raise_if_missing\u001B[49m\u001B[43m(\u001B[49m\u001B[43mkeyarr\u001B[49m\u001B[43m,\u001B[49m\u001B[43m \u001B[49m\u001B[43mindexer\u001B[49m\u001B[43m,\u001B[49m\u001B[43m \u001B[49m\u001B[43maxis_name\u001B[49m\u001B[43m)\u001B[49m\n\u001B[1;32m   6072\u001B[0m keyarr \u001B[38;5;241m=\u001B[39m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mtake(indexer)\n\u001B[1;32m   6073\u001B[0m \u001B[38;5;28;01mif\u001B[39;00m \u001B[38;5;28misinstance\u001B[39m(key, Index):\n\u001B[1;32m   6074\u001B[0m     \u001B[38;5;66;03m# GH 42790 - Preserve name from an Index\u001B[39;00m\n",
      "File \u001B[0;32m/mnt/Hossein-HDD/Projects/Fake-news-detection/venv/lib/python3.11/site-packages/pandas/core/indexes/base.py:6133\u001B[0m, in \u001B[0;36mIndex._raise_if_missing\u001B[0;34m(self, key, indexer, axis_name)\u001B[0m\n\u001B[1;32m   6130\u001B[0m     \u001B[38;5;28;01mraise\u001B[39;00m \u001B[38;5;167;01mKeyError\u001B[39;00m(\u001B[38;5;124mf\u001B[39m\u001B[38;5;124m\"\u001B[39m\u001B[38;5;124mNone of [\u001B[39m\u001B[38;5;132;01m{\u001B[39;00mkey\u001B[38;5;132;01m}\u001B[39;00m\u001B[38;5;124m] are in the [\u001B[39m\u001B[38;5;132;01m{\u001B[39;00maxis_name\u001B[38;5;132;01m}\u001B[39;00m\u001B[38;5;124m]\u001B[39m\u001B[38;5;124m\"\u001B[39m)\n\u001B[1;32m   6132\u001B[0m not_found \u001B[38;5;241m=\u001B[39m \u001B[38;5;28mlist\u001B[39m(ensure_index(key)[missing_mask\u001B[38;5;241m.\u001B[39mnonzero()[\u001B[38;5;241m0\u001B[39m]]\u001B[38;5;241m.\u001B[39munique())\n\u001B[0;32m-> 6133\u001B[0m \u001B[38;5;28;01mraise\u001B[39;00m \u001B[38;5;167;01mKeyError\u001B[39;00m(\u001B[38;5;124mf\u001B[39m\u001B[38;5;124m\"\u001B[39m\u001B[38;5;132;01m{\u001B[39;00mnot_found\u001B[38;5;132;01m}\u001B[39;00m\u001B[38;5;124m not in index\u001B[39m\u001B[38;5;124m\"\u001B[39m)\n",
      "\u001B[0;31mKeyError\u001B[0m: \"['combined'] not in index\""
     ]
    }
   ],
   "source": [
    "# Finalize the model (optional)\n",
    "final_model = finalize_model(tuned_rf)\n",
    "\n",
    "# Predictions on the test set\n",
    "test_predictions = predict_model(tuned_rf, data=test)\n",
    "\n",
    "# Display the predictions\n",
    "print(test_predictions.head())"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-02-05T20:16:44.392972165Z",
     "start_time": "2024-02-05T20:16:38.846484501Z"
    }
   },
   "id": "e11f290ffc07a724"
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [],
   "metadata": {
    "collapsed": false
   },
   "id": "12774c1ca49f9e82"
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}